Abstract Technical organizations increasingly rely on innovation contests to find novel ideas for designing complex systems. These activities involve outsiders in the early stages of the design process, potentially leading to ground-breaking designs that surpass expectations. Here, the contest's rules document plays a crucial role: this design artifact communicates the organization's problem and the desired system performance to the participants—significantly impacting the resulting solutions. However, the contest's nature amplifies the challenges of communicating complex design problems across boundaries. Existing strategies for formulating—i.e., requirement and objective allocation—might not suit this context. We developed an inductive model of their formulation process based on a multi-year field study of five complex innovation contests. We found that the formulation team (or “seeker”) balanced the need to communicate their problem in detail with the risk of excluding valuable participants. Here, they chose among three approaches—incentivize, impose, or subsume—depending on their knowledge of potential solutions and the participants' capabilities. Notably, the seeker formulated more granularly than the literature describes, employing multiple approaches within each rules document. These findings shed light on a poorly understood aspect of innovation contests, resolve a longstanding debate in the engineering design literature, and guide practitioners' formulation processes.
{"title":"REQUIREMENTS, OBJECTIVES, BOTH, OR NEITHER: HOW TO FORMULATE COMPLEX DESIGN PROBLEMS FOR INNOVATION CONTESTS","authors":"Ademir-Paolo Vrolijk, Zoe Szajnfarber","doi":"10.1115/1.4063568","DOIUrl":"https://doi.org/10.1115/1.4063568","url":null,"abstract":"Abstract Technical organizations increasingly rely on innovation contests to find novel ideas for designing complex systems. These activities involve outsiders in the early stages of the design process, potentially leading to ground-breaking designs that surpass expectations. Here, the contest's rules document plays a crucial role: this design artifact communicates the organization's problem and the desired system performance to the participants—significantly impacting the resulting solutions. However, the contest's nature amplifies the challenges of communicating complex design problems across boundaries. Existing strategies for formulating—i.e., requirement and objective allocation—might not suit this context. We developed an inductive model of their formulation process based on a multi-year field study of five complex innovation contests. We found that the formulation team (or “seeker”) balanced the need to communicate their problem in detail with the risk of excluding valuable participants. Here, they chose among three approaches—incentivize, impose, or subsume—depending on their knowledge of potential solutions and the participants' capabilities. Notably, the seeker formulated more granularly than the literature describes, employing multiple approaches within each rules document. These findings shed light on a poorly understood aspect of innovation contests, resolve a longstanding debate in the engineering design literature, and guide practitioners' formulation processes.","PeriodicalId":50137,"journal":{"name":"Journal of Mechanical Design","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135344831","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract Design artifacts provide a mechanism for illustrating design information and concepts, but their effectiveness relies on alignment across design agents in what these artifacts represent. This work investigates the agreement between multi-modal representations of design artifacts by humans and artificial intelligence (AI). Design artifacts are considered to constitute stimuli designers interact with to become inspired (i.e., inspirational stimuli), for which retrieval often relies on computational methods using AI. To facilitate this process for multi-modal stimuli, a better understanding of human perspectives of non-semantic representations of design information, e.g., by form or function-based features, is motivated. This work compares and evaluates human and AI-based representations of 3D-model parts by visual and functional features. Humans and AI were found to share consistent representations of visual and functional similarities, which aligned well to coarse, but not more granular, levels of similarity. Human-AI alignment was higher for identifying low compared to high similarity parts, suggesting mutual representation of features underlying more obvious than nuanced differences. Human evaluation of part relationships in terms of belonging to same or different categories revealed that human and AI-derived relationships similarly reflect concepts of “near” and “far”. However, levels of similarity corresponding to “near” and “far” differed depending on the criteria evaluated, where “far” was associated with nearer visually than functionally related stimuli. These findings contribute to a fundamental understanding of human evaluation of information conveyed by AI-represented design artifacts needed for successful human-AI collaboration in design.
{"title":"Comparing and evaluating human and computationally derived representations of non-semantic design information","authors":"Elisa Kwon, Kosa Goucher-Lambert","doi":"10.1115/1.4063567","DOIUrl":"https://doi.org/10.1115/1.4063567","url":null,"abstract":"Abstract Design artifacts provide a mechanism for illustrating design information and concepts, but their effectiveness relies on alignment across design agents in what these artifacts represent. This work investigates the agreement between multi-modal representations of design artifacts by humans and artificial intelligence (AI). Design artifacts are considered to constitute stimuli designers interact with to become inspired (i.e., inspirational stimuli), for which retrieval often relies on computational methods using AI. To facilitate this process for multi-modal stimuli, a better understanding of human perspectives of non-semantic representations of design information, e.g., by form or function-based features, is motivated. This work compares and evaluates human and AI-based representations of 3D-model parts by visual and functional features. Humans and AI were found to share consistent representations of visual and functional similarities, which aligned well to coarse, but not more granular, levels of similarity. Human-AI alignment was higher for identifying low compared to high similarity parts, suggesting mutual representation of features underlying more obvious than nuanced differences. Human evaluation of part relationships in terms of belonging to same or different categories revealed that human and AI-derived relationships similarly reflect concepts of “near” and “far”. However, levels of similarity corresponding to “near” and “far” differed depending on the criteria evaluated, where “far” was associated with nearer visually than functionally related stimuli. These findings contribute to a fundamental understanding of human evaluation of information conveyed by AI-represented design artifacts needed for successful human-AI collaboration in design.","PeriodicalId":50137,"journal":{"name":"Journal of Mechanical Design","volume":"133 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135344962","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Flexure-based grippers offer an attractive alternative to conventional grippers used in robotics and automation. However, most existing designs appear to suffer from insufficient range of motion, loadability and support stiffness. This paper presents an approach to obtain well-performing flexure hinges for compact anthropomorphic grippers made via metal additive manufacturing. We propose a flexure hinge architecture that achieves a high range of motion despite the challenging combination of a small design space, high Young's modulus and limited minimum feature size. Furthermore, we present an optimization procedure to generate suitable tendon-driven designs with high loadability. Using this framework, a flexure hinge with an outer diameter of 21.5 mm and range of motion of ±30 deg is synthesized. For the range of 0 to 30 deg simulations show a lateral loadability of 52.5 to 18.6 N and lateral support stiffness of 12309 to 11130 N/m, determined at a gripper interface located 41.2 mm from the hinge pivot axis. Experiments confirm a loadability of at least 15.4 N and determined a stiffness of 8982 to 9727 N/m for same conditions. The results show that the flexure hinge architecture has large potential for a wide range of applications, while in combination with the optimization procedure superior designs for tendon-driven grippers can be obtained.
{"title":"Flexure hinge design and optimization for compact anthropomorphic grippers made via metal additive manufacturing","authors":"M. Tschiersky, Jan De Jong, Dannis Brouwer","doi":"10.1115/1.4063362","DOIUrl":"https://doi.org/10.1115/1.4063362","url":null,"abstract":"\u0000 Flexure-based grippers offer an attractive alternative to conventional grippers used in robotics and automation. However, most existing designs appear to suffer from insufficient range of motion, loadability and support stiffness. This paper presents an approach to obtain well-performing flexure hinges for compact anthropomorphic grippers made via metal additive manufacturing. We propose a flexure hinge architecture that achieves a high range of motion despite the challenging combination of a small design space, high Young's modulus and limited minimum feature size. Furthermore, we present an optimization procedure to generate suitable tendon-driven designs with high loadability. Using this framework, a flexure hinge with an outer diameter of 21.5 mm and range of motion of ±30 deg is synthesized. For the range of 0 to 30 deg simulations show a lateral loadability of 52.5 to 18.6 N and lateral support stiffness of 12309 to 11130 N/m, determined at a gripper interface located 41.2 mm from the hinge pivot axis. Experiments confirm a loadability of at least 15.4 N and determined a stiffness of 8982 to 9727 N/m for same conditions. The results show that the flexure hinge architecture has large potential for a wide range of applications, while in combination with the optimization procedure superior designs for tendon-driven grippers can be obtained.","PeriodicalId":50137,"journal":{"name":"Journal of Mechanical Design","volume":"28 1","pages":""},"PeriodicalIF":3.3,"publicationDate":"2023-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75415835","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Many design problems involve reasoning about points in high-dimensional space. A common strategy is to first embed these high-dimensional points into a low-dimensional latent space. We propose that a good embedding should be isometric---i.e., preserving the geodesic distance between points on the data manifold in the latent space. However, enforcing isometry is non-trivial for common Neural embedding models such as autoencoders. Moreover, while theoretically appealing, it is unclear to what extent is enforcing isometry necessary for a given design analysis. This paper answers these questions by constructing an isometric embedding via an isometric autoencoder, which we employ to analyze an inverse airfoil design problem. Specifically, the paper describes how to train an isometric autoencoder and demonstrates its usefulness compared to non-isometric autoencoders on the UIUC airfoil dataset. Our ablation study illustrates that enforcing isometry is necessary for accurately discovering clusters through the latent space. We also show how isometric autoencoders can uncover pathologies in typical gradient-based Shape Optimization solvers through an analysis on the SU2-optimized airfoil dataset, wherein we find an over-reliance of the gradient solver on angle of attack. Overall, this paper motivates the use of isometry constraints in Neural embedding models, particularly in cases where researchers or designers intend to use distance-based analysis measures to analyze designs within the latent space. While this work focuses on airfoil design as an illustrative example, it applies to any domain where analyzing isometric design or data embeddings would be useful.
{"title":"Characterizing Designs via Isometric Embeddings: Applications to Airfoil Inverse Design","authors":"Qiuyi Chen, M. Fuge","doi":"10.1115/1.4063363","DOIUrl":"https://doi.org/10.1115/1.4063363","url":null,"abstract":"\u0000 Many design problems involve reasoning about points in high-dimensional space. A common strategy is to first embed these high-dimensional points into a low-dimensional latent space. We propose that a good embedding should be isometric---i.e., preserving the geodesic distance between points on the data manifold in the latent space. However, enforcing isometry is non-trivial for common Neural embedding models such as autoencoders. Moreover, while theoretically appealing, it is unclear to what extent is enforcing isometry necessary for a given design analysis. This paper answers these questions by constructing an isometric embedding via an isometric autoencoder, which we employ to analyze an inverse airfoil design problem. Specifically, the paper describes how to train an isometric autoencoder and demonstrates its usefulness compared to non-isometric autoencoders on the UIUC airfoil dataset. Our ablation study illustrates that enforcing isometry is necessary for accurately discovering clusters through the latent space. We also show how isometric autoencoders can uncover pathologies in typical gradient-based Shape Optimization solvers through an analysis on the SU2-optimized airfoil dataset, wherein we find an over-reliance of the gradient solver on angle of attack. Overall, this paper motivates the use of isometry constraints in Neural embedding models, particularly in cases where researchers or designers intend to use distance-based analysis measures to analyze designs within the latent space. While this work focuses on airfoil design as an illustrative example, it applies to any domain where analyzing isometric design or data embeddings would be useful.","PeriodicalId":50137,"journal":{"name":"Journal of Mechanical Design","volume":"31 1","pages":""},"PeriodicalIF":3.3,"publicationDate":"2023-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73127229","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents a 3D material mask overlay topology optimization approach using truncated octahedron elements and spheroidal masks. Truncated octahedron elements provide face connectivity between two juxtaposed elements, thus, eliminating singular solutions inherently. A novel meshing scheme with Tetra-Kai-Decaheral or TKD (generic case of truncated octahedron) elements is proposed. The scheme is extended to parameterized generic-shape domains. Various benefits of implementing the elements are also highlighted, and the corresponding finite element is introduced. Spheroidal negative masks are employed to determine the material within the elements. Seven design variables define each mask. A material density formulation is proposed, and sensitivity analysis for gradient-based optimization is developed. fmincon MATLAB function is used for the optimization. The efficacy and success of the approach are demonstrated by solving structures and compliant mechanism design problems. Compliance is minimized for the former, whereas a multi-criteria arising due to flexibility and stiffness measures is extremized for optimizing the mechanisms. Convergence of the optimization is smooth. The volume constraint is satisfied and remains active at the end of the optimization.
{"title":"3D material mask overlay topology optimization approach with truncated-octahedron elements","authors":"Nikhil Singh, Prabhat Kumar, A. Saxena","doi":"10.1115/1.4063361","DOIUrl":"https://doi.org/10.1115/1.4063361","url":null,"abstract":"\u0000 This paper presents a 3D material mask overlay topology optimization approach using truncated octahedron elements and spheroidal masks. Truncated octahedron elements provide face connectivity between two juxtaposed elements, thus, eliminating singular solutions inherently. A novel meshing scheme with Tetra-Kai-Decaheral or TKD (generic case of truncated octahedron) elements is proposed. The scheme is extended to parameterized generic-shape domains. Various benefits of implementing the elements are also highlighted, and the corresponding finite element is introduced. Spheroidal negative masks are employed to determine the material within the elements. Seven design variables define each mask. A material density formulation is proposed, and sensitivity analysis for gradient-based optimization is developed. fmincon MATLAB function is used for the optimization. The efficacy and success of the approach are demonstrated by solving structures and compliant mechanism design problems. Compliance is minimized for the former, whereas a multi-criteria arising due to flexibility and stiffness measures is extremized for optimizing the mechanisms. Convergence of the optimization is smooth. The volume constraint is satisfied and remains active at the end of the optimization.","PeriodicalId":50137,"journal":{"name":"Journal of Mechanical Design","volume":"45 1","pages":""},"PeriodicalIF":3.3,"publicationDate":"2023-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82134090","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
As computer-aided design (CAD) tools have become an essential aspect of modern mechanical engineering design, the demand for CAD experts has increased significantly. The development from novice, to proficient, to expert user is of particular interest to the industrial and academic design communities. Yet little is known about the development or characteristics of expert CAD skill; much of the past work that reports user action data is based on student or novice data. We compared the CAD modelling process across nine proficient and ten expert designers as they were tested to complete the same design task. Under identical conditions – the same time constraints in the same CAD platform and with the same task -- the expert users were able to complete a larger proportion of the task with higher dimensional accuracy. While the experts were able to dissect and retrieve geometries from manufacturing drawings more efficiently than proficient users, they were also able to plan a modelling strategy that required less effort and revisions. With our experimental findings, we identify the demand for procedural knowledge-building for young engineers, with the ultimate goal of more effectively developing experts in engineering design with CAD.
{"title":"What sets proficient and expert users apart? Results of a Computer-Aided Design experiment","authors":"Yuan Deng, James Chen, A. Olechowski","doi":"10.1115/1.4063360","DOIUrl":"https://doi.org/10.1115/1.4063360","url":null,"abstract":"\u0000 As computer-aided design (CAD) tools have become an essential aspect of modern mechanical engineering design, the demand for CAD experts has increased significantly. The development from novice, to proficient, to expert user is of particular interest to the industrial and academic design communities. Yet little is known about the development or characteristics of expert CAD skill; much of the past work that reports user action data is based on student or novice data. We compared the CAD modelling process across nine proficient and ten expert designers as they were tested to complete the same design task. Under identical conditions – the same time constraints in the same CAD platform and with the same task -- the expert users were able to complete a larger proportion of the task with higher dimensional accuracy. While the experts were able to dissect and retrieve geometries from manufacturing drawings more efficiently than proficient users, they were also able to plan a modelling strategy that required less effort and revisions. With our experimental findings, we identify the demand for procedural knowledge-building for young engineers, with the ultimate goal of more effectively developing experts in engineering design with CAD.","PeriodicalId":50137,"journal":{"name":"Journal of Mechanical Design","volume":"30 1","pages":""},"PeriodicalIF":3.3,"publicationDate":"2023-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74308101","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Delta-like architectures are widely used for fast pick-and-place applications. When rotational degrees of freedom are required to perform a task, one or more UPU kinematic chains are usually added to transmit the torques from motors located on the base to the platform, in order to actuate a wrist. Packaging applications usually require five degrees of freedom, and two UPU chains are then used to actuate two rotational degrees-of-freedom (DOFs) on the end-effector. However, the UPU chain induces significant limitations for industrial use: it significantly constrains the workspace along the vertical direction and implies a backlash in the universal joints degrading the accuracy of the robot. In this paper, we investigate an alternative to the UPU kinematic chain for designing Delta-like robots with five DOFs. Indeed, the actuation of a two-DOFs wrist is performed through the use of a kinematic chain based on a succession of parallelograms associated with a Delta-like leg. After a description of the kinematic models of the modified leg and an analysis of its singularities, a design optimization procedure is presented in order to define suitable geometric parameters for a given industrial application. Finally, a prototype is presented and its performances are evaluated.
{"title":"Modeling and Design of a five Degrees-of-Freedom Delta-Like Robot for Fast Pick-and-Place Applications","authors":"Valentin Le Mesle, Vincent Bégoc, S. Briot","doi":"10.1115/1.4063359","DOIUrl":"https://doi.org/10.1115/1.4063359","url":null,"abstract":"\u0000 Delta-like architectures are widely used for fast pick-and-place applications. When rotational degrees of freedom are required to perform a task, one or more UPU kinematic chains are usually added to transmit the torques from motors located on the base to the platform, in order to actuate a wrist. Packaging applications usually require five degrees of freedom, and two UPU chains are then used to actuate two rotational degrees-of-freedom (DOFs) on the end-effector. However, the UPU chain induces significant limitations for industrial use: it significantly constrains the workspace along the vertical direction and implies a backlash in the universal joints degrading the accuracy of the robot. In this paper, we investigate an alternative to the UPU kinematic chain for designing Delta-like robots with five DOFs. Indeed, the actuation of a two-DOFs wrist is performed through the use of a kinematic chain based on a succession of parallelograms associated with a Delta-like leg. After a description of the kinematic models of the modified leg and an analysis of its singularities, a design optimization procedure is presented in order to define suitable geometric parameters for a given industrial application. Finally, a prototype is presented and its performances are evaluated.","PeriodicalId":50137,"journal":{"name":"Journal of Mechanical Design","volume":"151 1","pages":""},"PeriodicalIF":3.3,"publicationDate":"2023-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73460428","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Power-split hybrid transmissions are the core components of power-split hybrid electric vehicles (HEV), and the quest for a more energy-efficient and higher-performing power-split hybrid transmission has long been the focus of study. In contrast to previously published methodologies, this paper proposes a novel approach for directly synthesizing power-split hybrid transmissions that makes use of the results of previously synthesized planetary gear trains (PGTs) rather than necessitating a re-synthesis of their PGTs. A new topological graph that can construct a bridge between the PGTs and power-split hybrid transmission has been developed, reducing the computational complexity of the synthesis process. The new topological graph is obtained by adding the topological characteristics of the power-split hybrid transmission to the PGT graph. A standard structure matrix is proposed to further screen out all the isomorphic configurations. The present method can generate various types of multi-PGT hybrid transmissions while avoiding mechanical and structural interference. The design process of configurations for power-split hybrid transmission with 3-column PGTs (3-PGT) is used as an example to prove the rationality of the method.
{"title":"Topological Graph Representation and Configuration Synthesis for Power split Hybrid Transmissions of Multi-Planetary Gear Trains","authors":"Meijie Geng, H. Ding, Tao Ke, Wenjian Yang","doi":"10.1115/1.4063287","DOIUrl":"https://doi.org/10.1115/1.4063287","url":null,"abstract":"\u0000 Power-split hybrid transmissions are the core components of power-split hybrid electric vehicles (HEV), and the quest for a more energy-efficient and higher-performing power-split hybrid transmission has long been the focus of study. In contrast to previously published methodologies, this paper proposes a novel approach for directly synthesizing power-split hybrid transmissions that makes use of the results of previously synthesized planetary gear trains (PGTs) rather than necessitating a re-synthesis of their PGTs. A new topological graph that can construct a bridge between the PGTs and power-split hybrid transmission has been developed, reducing the computational complexity of the synthesis process. The new topological graph is obtained by adding the topological characteristics of the power-split hybrid transmission to the PGT graph. A standard structure matrix is proposed to further screen out all the isomorphic configurations. The present method can generate various types of multi-PGT hybrid transmissions while avoiding mechanical and structural interference. The design process of configurations for power-split hybrid transmission with 3-column PGTs (3-PGT) is used as an example to prove the rationality of the method.","PeriodicalId":50137,"journal":{"name":"Journal of Mechanical Design","volume":"107 1","pages":""},"PeriodicalIF":3.3,"publicationDate":"2023-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79294093","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Recently, online user-generated data has emerged as a valuable source for consumer product research. However, most studies have neglected the brand effect, although it is a significant factor in conventional market research. This paper demonstrates the importance of brands in data-driven design using online reviews. Specifically, this study utilizes game theory and suggests a game setting representing market competition. Elements of the game are determined based on online data analysis. The proposed approach consists of three stages. The first stage divides online customers into different segments and analyzes them to extract the feature importance of each brand in each segment. The importance is based on the positive term frequency of features, and it becomes the customer’s partial utility for each feature. The second stage defines the specification of product candidates and calculates their costs. This study refers to real market datasets (Bill of Materials) available online. At this point, the game is all set. The final stage finds the Nash Equilibrium of the designed game and compares the optimal strategy for a product portfolio with and without brand consideration. The suggested approach was tested on smartphone reviews from Amazon. The result shows that the lack of brand consideration leads a company to choose a non-optimal product strategy, illustrating the significance of the brand factor. Keywords: data-driven design, online review, brand effect
{"title":"Analysis of Brand Effects in Data-Driven Design Based on Online Reviews","authors":"Seyoung Park, Harrison M. Kim","doi":"10.1115/1.4063288","DOIUrl":"https://doi.org/10.1115/1.4063288","url":null,"abstract":"\u0000 Recently, online user-generated data has emerged as a valuable source for consumer product research. However, most studies have neglected the brand effect, although it is a significant factor in conventional market research. This paper demonstrates the importance of brands in data-driven design using online reviews. Specifically, this study utilizes game theory and suggests a game setting representing market competition. Elements of the game are determined based on online data analysis. The proposed approach consists of three stages. The first stage divides online customers into different segments and analyzes them to extract the feature importance of each brand in each segment. The importance is based on the positive term frequency of features, and it becomes the customer’s partial utility for each feature. The second stage defines the specification of product candidates and calculates their costs. This study refers to real market datasets (Bill of Materials) available online. At this point, the game is all set. The final stage finds the Nash Equilibrium of the designed game and compares the optimal strategy for a product portfolio with and without brand consideration. The suggested approach was tested on smartphone reviews from Amazon. The result shows that the lack of brand consideration leads a company to choose a non-optimal product strategy, illustrating the significance of the brand factor. Keywords: data-driven design, online review, brand effect","PeriodicalId":50137,"journal":{"name":"Journal of Mechanical Design","volume":"90 1","pages":""},"PeriodicalIF":3.3,"publicationDate":"2023-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76337955","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Christopher S. Mabey, Erin Peiffer, Nordica A. MacCarty, Christopher A. Mattson
This paper presents a methodology for predicting the adoption and social impact of a product using agent-based modeling (ABM) and neural networks to aid in decision-making related to the design and implementation of the product in a sociotechnical system. The collection of primary data on the social impact of a product is also outlined. Although this paper illustrates the method for improved cookstoves in Uganda, the method can be applied to a wide range of contexts. A field study was carried out in Uganda, consisting of two phases of data collection. The data from the fieldwork was used to train a neural network to predict if an individual would adopt an improved cookstove. Data collected from surveys and the trained adoption model were used to create an ABM to estimate adoption rates and social impacts experienced by households that had adopted technology and to assess social impact indicators. The contributions of this article are a method for collecting primary social impact data on a product and how to integrate those data into a predictive agent-based social impact model. This methodology also enables the examination of leverage points in the sociotechnical system to improve the social impact of a product as it is implemented in society.
{"title":"Simulating the Adoption and Social Impact of Improved Cookstoves in Uganda Using Agent-Based Modeling and Neural Networks","authors":"Christopher S. Mabey, Erin Peiffer, Nordica A. MacCarty, Christopher A. Mattson","doi":"10.1115/1.4063237","DOIUrl":"https://doi.org/10.1115/1.4063237","url":null,"abstract":"\u0000 This paper presents a methodology for predicting the adoption and social impact of a product using agent-based modeling (ABM) and neural networks to aid in decision-making related to the design and implementation of the product in a sociotechnical system. The collection of primary data on the social impact of a product is also outlined. Although this paper illustrates the method for improved cookstoves in Uganda, the method can be applied to a wide range of contexts. A field study was carried out in Uganda, consisting of two phases of data collection. The data from the fieldwork was used to train a neural network to predict if an individual would adopt an improved cookstove. Data collected from surveys and the trained adoption model were used to create an ABM to estimate adoption rates and social impacts experienced by households that had adopted technology and to assess social impact indicators. The contributions of this article are a method for collecting primary social impact data on a product and how to integrate those data into a predictive agent-based social impact model. This methodology also enables the examination of leverage points in the sociotechnical system to improve the social impact of a product as it is implemented in society.","PeriodicalId":50137,"journal":{"name":"Journal of Mechanical Design","volume":"45 1","pages":""},"PeriodicalIF":3.3,"publicationDate":"2023-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73810242","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：